cond

Condition number for inversion.

Syntax

  • c = rcond(A, p)

Input argument

  • A - a numeric value: square or rectangular (double or single)

  • p - norm type: Inf, 'fro', 1, 2 (default)

Output argument

  • c - a numeric value: a scalar.

Description

c = cond(A) returns the 2-norm condition number for inversion.

c = cond(A, p) returns the p-norm condition number, where p can be 1, 2, Inf, or 'fro'.

Example

X = rand(10, 10);
r = cond(X)

See also

rcond.

History

VersionDescription

1.0.0

initial version

Author

Allan CORNET

Last updated